Suicide, the second leading cause of death in adolescents, is the most tragic complication of a psychiatric condition in this age group. Every year, approximately 157,000 youth receive medical care for suicide related injuries at emergency departments throughout the U.S. Despite some progress, suicide prevention continues to be a daunting task. In adolescents, the risk of a second suicide attempt is approximately 30% after discharge from an inpatient psychiatric unit. Up to 80% of suicidal patients who subsequently died by suicide deny suicidal ideation in their last communication with a health care provider. Therefore, there is an urgent need for the development of biomarkers that can objectively identify which youth are most likely to engage in subsequent suicide attempts. Several lines of evidence (postmortem studies, genetic studies, biomarker studies) as well as preliminary studies conducted by our group have pointed to neuroinflammation as one of the neurobiological findings observed in suicidal behavior. In particular, the principal investigator and co investigators have identified S1OOB - an astrocytic protein, which is a marker of blood brain barrier (BBB) impairment, as a novel biomarker associated with suicidality in adolescents. We are now also investigating two additional and important markers Kynureninic Acid (KYNA) and Quinolinic Acid (QUIN) identified by Dr. Brundin (co-investigator) to be altered in patients after a suicide attempt. Other studies have reported several other peripheral inflammatory markers (PlMs) including IL-1, IL- 6 and TNF-? as associated with suicidality. Hence, PlMs either on their own or along with clinical markers may be particularly useful in predicting future suicide attempts. This study will investigate whether PIM levels, with or without clinical predictors, can be useful at the time of discharge from an inpatient psychiatric unit to predict suicidality in adolescent patients in the subsequent 12 months. The first aim of this study is to determine whether plasma levels of S100B, IL-1, IL-6, TNF-?, KYNA, and QUIN correlate with suicidal behavior (SB) .Secondly, this study will investigate if any of the PlMs can predict suicidal attempts. Finally, we will test which combination of clinical risk factors and PlMs is able to most efficiently predict SB post-discharge from the inpatient unit. Innovative aspects of this proposal include: 1) The first study to longitudinally investigate levels of PlMs at the time of admission and their change at the time of discharge in adolescent patients being admitted for SB. 2) The first study to investigate whether level of PIMS (alone or in combination with clinical risk factors) at the time of discharge can predict readmissions for SB in the next 12 months. 3) Beside the well-studied PIMs in adult samples, specifically investigate novel biomarkers of inflammation- S100B and 2 markers of the Kynurenine pathway.